UEG's educational platform for gastroenterology education bringing recordings, courses, guidelines, and more together in one place.

UEG Week Berlin 2025 Recordings Gut Guide online

Rate for better recommendations!

Your feedback helps us show you more of what you like and less of what you dislike!

More of this!

We will show you more similar content.

Not for me!

We won't suggest this again.

Bookmark!

Save this content for later!

New to myUEG? Create an account! Log In with myUEG account

UEG Presentation

Share via Email Share on Facebook Share on X Share on LinkedIn Share on Bluesky

VALIDATION OF A BILIOPANCREATIC EUS-INTELLIGENT AND REAL-TIME ENDOSCOPY ANALYTICAL DEVICE FOR REAL-TIME QUALITY IMPROVEMENT: A PROPECTIVE, SINGLE-CENTER, RANDOMIZED CONTROLLED TRIAL

Free for all myUEG account holders. Your access level is set automatically based on your occupation. Medical professionals get full access to all content. If you are a non-medical user, you can only access UEG Week content from congresses you attended.

Not sure what you can access? Learn more about account types.

Introduction

Endoscopic ultrasonography (EUS) is a key procedure for diagnosing biliopancreatic diseases. However, the performance among EUS endoscopists varies greatly and leads to blind spots during operation, which impaired the health outcome of patients. Comprehensive scanning without blind area is crucial for optimising detection rate. However, two factors lead to the blind spots of biliopancreatic EUS procedure: Firstly, the absence of standarzation. To standardize EUS procedures, a multi-station approach in biliopancreatic EUS has been established as the standard according to the Japanese Society of Gastrointestinal Endoscopy and European and American experts. However, they are often not well followed from lack of supervision and the availability of practical tools. Secondly, it is difficult to understand the ultrasound images. It is challenging for endoscopists to understand the ultrasound images because of the subtleties of interpretation, small observation fields, uncertainty in transducer position and orientation, etc. Therefore, it is critical to develop a practical tool that can monitor the blind spots and reduce the difficulty of ultrasound image interpretation in real-time to standardize the quality of EUS procedure. We previously developed an artificial intelligence device that accurately identifies EUS standard stations and significantly reduces the difficulty of ultrasound image interpretation, but the validation in clinical practice is still lacking. In this study, we updated the device (named EUS-IREAD) and assessed its performance in improving the quality of EUS procedure in a single-center randomized controlled trial.

Aims & Methods

The device consisted of five 5 learning models to identify 8 EUS stations (Station 1: Abdominal aorta, Station 2: Pancreatic body, Station 3: Pancreatic tail, Station 4: Confluence, Station 5: Liver hilum. Station 6: Pancreatic head from stomach. Station 7: Duodenal bulb, Station 8: Descending duodenum) and 24 anatomic structures. Patients were recruited from Renmin Hospital of Wuhan University from July 09, 2022, to February 28, 2023, and randomly assigned to undergo the EUS procedure with or without the assistance of EUS-IREAD. The primary outcome was the missed scanning rate of standard stations between the two groups.

Results

290 patients with suspected biliopancreatic diseases were randomly recruited and analyzed, including 144 in the EUS-IREAD-assisted group and 146 in the control group. Comparing with the control group, the EUS-IREAD-assisted group has lower missed scanning rate of stations (4.47% [SD 0.78] vs 14.27% [SD 0.95]; -9.81%, 95%Cl, [-0.12 - -0.07]; P<0.001) and lower missed scanning rate of anatomical structures (6.2% [SD 2.75] vs 17.39% [SD 3.17]; -11.19%, 95%Cl, [-0.13 - -0.1]; P<0.001) (Table).

point	Mean (SD, median, IQR)		Difference (95% CI)	P value
point	EUS-IREAD-assisted group (N=144)	Control group (N=146)	Difference (95% CI)	P value
The missed scanning rate of standard stations (%)	4·5(0·78, 8, 1)	14·3(0·95, 7, 2)	-9·8% (-12·2%, -7·5%)	<0·001
Gastric sites	2·8(0·44, 6, 0)	12·1(0·81, 5, 1)	-9·3% (-11·7%, -6·9 %)	<0·001
Duodenal sites	9·7(0·6, 2, 0)	21·1(0·74, 2, 1)	-11·4% (-17·3%, -5·4%)	<0·001
The missed scanning rate of anatomical structures (%)	6·2(2·75, 24, 3)	17·4(3·17, 20, 6)	-11·2% (-12·7%, -9·7%)	<0·001
Pancreas	4·5(0·7, 6, 0)	12·9(0·84, 5, 1)	-8·4% (-11%, -5·8%)	<0·001
Blood vessels	4·8(0·79, 8, 1)	14·5(1·08, 7, 2)	-9·7% (-12·1%, -7·3%)	<0·001
Biliopancreatic duct	10(1·25, 7, 1)	24·2(1·42, 5, 2)	-14·2% (-17·5%, -11%)	<0·001
Kidney&spleen	3·5(0·28, 2, 0)	16·1(0·61, 2, 0)	-12·6% (-17·3%, -7·9%)	<0·001

Conclusion

Our study confirm its capability to monitor the blind spots and reduce the missed rate of stations and structures in EUS procedure. The EUS-IREAD has the potential to play an essential role in EUS quality control.